Evaluation of hospital factors associated with hospital postoperative venous thromboembolism imaging utilisation practices

Background: Recent research suggests that hospital rates of postoperative venous thromboembolism (VTE) are subject to surveillance bias: the more hospitals 'look for' VTE, the more VTE they find. However, little is known about what drives variation in hospital VTE imaging rates. We conducted an observational study to examine hospital and market characteristics that were associated with hospital-level rates of postoperative VTE imaging, focusing on hospitals with particularly high rates.

Methods: For Medicare beneficiaries undergoing 11 major operations (2009-2010) at 2820 hospitals, hospital-level postoperative VTE imaging use rates were calculated. Hospital characteristics associated with hospital VTE imaging use rates were examined including case severity, size, ownership, VTE process measure adherence, accreditations, staffing, malpractice environment, and county market factors. Associations between explanatory variables and VTE imaging rates were assessed using quantile regressions at the 25th, median, 75th and 90th quantiles.

Results: Mean postoperative VTE imaging rates ranged from 85.26 (SD=67.38) per 1000 discharges in the lowest quartile of hospitals ranked by VTE imaging rates to 168.86 (SD=76.70) in the highest quartile. Drivers of high imaging rates at the 90th quantile were high resident-to-bed ratio (coefficient=51.35, p<0.01), Joint Commission accreditation (coefficient=19.05, p<0.01), presence of other hospitals in the same market with high imaging rates (coefficient=15.29, p<0.01), average case severity (coefficient=11.97, p<0.01), local malpractice costs (coefficient=11.29, p<0.01), and market competition (coefficient=11.03, p<0.01).

Conclusions: Hospital teaching status, resident-to-bed ratio, malpractice environment and local market factors drive hospital postoperative VTE imaging use, suggesting that non-clinical forces predominantly drive hospital VTE imaging practices.